V O L U M E 21, NO. 11, N O V E M B E R 1 9 4 9 necessary to remove the oxides by filtration bpfore tiedment with formic acid. When the metal or alloy to be analyzed contains little or no reducible metal ions it is possible to use less hypophosphorous acid in the distillation mixture. On the other hand, in the analysis of metals such as molybdenum, vanadium, iron, copper, or tin the concentration recommended in the procedure must be used in order to assure rapid and complete reduction of the iodine formed. If the iodine is not reduced by the hypophosphorous acid before distillation takes place, some of it may be carried over into the ammonium hydroxide solution, resulting in oxidation of the sulfide when the latter is subsequently distilled over. If necessary from an economy standpoint, the amount of distillation mixture or the concentration of hydriodic acid therein may be reduced in certain analyses. On the other hand, reduction is more rapid with the recommended distillation mixture and in addition, the latter has been adjusted to enwre solution of 1 gram of lead, tin, or copper during distillation. EXPERIMENTAL
Typical calibration curve data, obtained as directed in the procedure, are recorded in Table I. Various metals and alloys were analyzed as directed in the procedure, using the appropriate modifications whenever necessary. Formic arid was removed in all the analyses except that of molybdenum (Table 11). In some cases the values given are an average of duplicate or triplicate analyses. In such instances the maximum spread in the analyses was of the order of 1 or 2 micrograms of sulfur. Because of the lack of suitable standards with which to “prove in” the method, it was necessary in many instances to resort to an indirect procedure for checking the method. This consisted of analyzing two samples of the metal or alloy, to one of which were added 25 micrograms of sulfur in the form of potassium sulfate. (The metal sample was added to a flask in which a 5-ml. aliquot of standard potassium sulfate had been evaporated nearly to dryness.) If the difference in the quantitv of sulfur obtained in the two analyses was close to 25 micrograms, it x i s assumed that the method was satisfactory.
1373 The percentages of sulfur found from the first analysis are listed in the nest to the last column of Table 11. The differences in quantity of sulfur obtained in the first and second analysis are listed in the last column of Table 11. In order to investigate the effect of formic acid, several metal samples were dissolved and distilled as directed in the procedure. Upon completion of the distillation the solution in the volumetric flask was quickly transferred to a clean absorption cell without adding lead citrate or diluting to 25 ml. and photometered at 370 mp. The process was repeated, with the exception that the excess formic acid was not boiled off. Instead, the solution was boiled down to 5 ml. after the destruction of the nitric arid. The results are recorded in Table 111.
Table 111. so. 1
’2
3 4 5 6 7 8
Effect of Formic Acid
Formic Acid Present %T B. of S.ingot iron No. 55b 98 A.S.T.M. electronic nickel No. H-1400 98 B. of S.zinc base No. 94 98 OFHC copper metal 96 Bismuth metal 96 B. of S. solder No. 127 98 B. of S.lead base KO,53b E5 89 Lead (as PbClz) Sample
Formic Acid Expelled % T 100 100 100 100 100
100 97 98
ACKNOWLEDGMENT
The author is indebted to L. -4.Wooten for critical comments, to Miss D. 31. Dodd for the spectrophotometric investigation of the lead sol, and to Mrs. M. H. Read for spectrochemical detection of metal ion contamination of the distillates. LITERATURE CITED (1) Field, E.. and Oldach, C . S.,IND. ENQ.CHEY.,-4h-a~. ED., 18, 665 (1946). 12) Luke, C. L., Ibid., 15, 602 (1943). (3) I b i d . , 17, 298 (1945). (4) Sherman, M . , Am. Foundryman, 13, KO.3 , 52 (1948). RECEIVED March 16, 1949.
Determination of Zinc Oxide as a Residual of Zinc Powder E. W. BALIS, L. B . BRONK, AND H. A. LIEBHAFSKY, General Electric Company, Schenectady, !V. Y .
A simple technique, which may be applicable in other cases, has been devised for determining the zinc oxide on powdered zinc. The powder was spread on a copper boat, covered, wrapped tightly in annealed copper foil, and heated in vacuum at 450” C. In this way, the zinc was completely removed and the zinc oxide could be brushed off and weighed on the micmbalance.
I
1946, this laboratory undertook to isolate, identify, and determine the zinc oxide present as a film ol fine, reagent grade zinc powder. This objective was accomplished by obtainine the zinc oxide as a residual after the metal had been completely removed by a combination of phi-sical processes carried out under conditions designed to minimize oxidation and loss of oxide by entrainment. The technique,emplog-ed is new and may prove useful in other cases. EXPERIMENTAL METHOD
The neighed sample of zinc powder (ca.,,200 mg.) is spread evenly on the center portion of the “boat, which is made by
splitting thin-walled (ca. 0.08 cm.) copper tubing (see Figure 1). After the boat has been covered with a similar section of tubing pressed into place, i t is tightly wrapped in annealed copper foil 0.003 cm, thick, the of which are folded over. B ~ fore beine used. all codoer is hGdronen-fired to remove oxide. UD to three \\-rapped b&b are then-stacked, and inserted in thk Pyrex firing tube, and the apparatus (Figure 2 ) is assembled for a run. The tube is flushed for 10 minutes with line hydrogen, evacuated rvith a good oil vacuum pump for 10 minutes, flushed with line hydrogen for an hour (less would probably suffice), and then reconnected to the pump. The temperature of the furnace is next raised to 450” C. and held there for 30 minutes to complete the separation of zinc and zinc oxide. Furnace and tube are now allowed to cool, whereupon air is readmitted. Bfter the boats have been removed and opened, the loosely adhering zinc oxide
ANALYTICAL CHEMISTRY
1374 Wrapped boat
carried out in high vacuum
t a n nigh vacuum m the simple apparatus of Figure 2. Ordinarily, this error will not be Figure 1. Boat a t Various Stages in an Experiment great enough to warrant the N o t e rino-rid pattern after B r i n ~ . Residual zinc oxide hap heen h r v a h d OUI use of the much mare elaborate high-vacuum equipment. Miscellaneous. The determination has heeu carried through with good results both %hove and below the melting point of zinc (419.4" C.), which .. .-. . . .. -. proves that removal of the metal need not in. ..-. ...- volve a liquid, (The copper-zinc diagram shows no phases melting below 419.4'C., 1 . ) Evapora0 2 4 6 INCHES LlllLu tion of einc and its diffusion into copper are inSCALE strumental in the process; even the outer wrapping of copper foil turns to brass. Figure 2. Apparatus with Thermocouple Added The ease and reliability with which the residual A. Thsrrnoooupl--not -"ked D. Wrapped hosts oxide can be brushed off the boat are striking. in usus1 snsbsir E. Famace B. Entry for hydrogen F. Exit oonnestion for On a 200-mg. sample, 0.01% of oxygen corresponds C. Pyraxtuhs .3raau**ion to about 0.1 mg. of zinc oxide. Price ( 5 ) desoribes the determination of zinc in its alloys by evaporation into vacuum and aives referenoes to powder is brushed onto a piece of platinum foil and weighed on 8. microbalance. earlier work. Vernon, Akeroyd, and Stroud ( 4 ) succeeded in Data. to show the reliability of th8 method are given in Table isolating oxide films from einc foil by the same method. I. Results thus obtained are slightly high. noat
t
*l",ple
DISCUSSION
The foregoing technique should be applicable to the isolation and determination of other oxides, and t o the analysis of alloys. The apparatus employed is available in most lahoratories. Shielding. During the preliminary work, in which the sample was shielded very little or not at all, rings of zino formed on the tube just beyond the ends of the furnace and zinc oxide was found on the tube between the sample and the rings of zinc. Moreover, too much zinc oxide generally remained in the boat, and in variable amount. atmosphere a t Because zinc has a vapor pressure near 400' C. (d), its evaporation under the experimental conditions is t o he expected. Zinc oxide, however, is much less volatile. The residues from two experiments lost no weight upon being heated for 1 and then far 2 hours under the experimental conditions. Although the zinc oxide ou the firing tube might possibly have been formed by the oxidation of evaporated metal, it is very much more likely to have been entrained by the vapor and then deposited on the tube. Effective shielding is therefore indicated. Such shielding would at the same time reduce the chance of stray oxidation during the run. Wrapping the covered boat in annealed, hydrogen-fired copper foil accomplished both objectives to a satisfactory extent, and prevented the formation of any visible deposit on the glass tube. Stray Oxidation. Table I gives evidence of stray oxidation, inasmuch 88 oxygen contents above 1.17% were found when heating wae prolonged. Determinations of Binc oxide were
Table I. Oxygen Content of Reagent Grade Zinc Powder Date
11/22/46 11/22/46
12/31/46
Minutes a t 450'C. (Furnace) 15 30 30
45 60 11/13/46 120 Zinopowder pB19ed 325-mesh sieYe,
% owgen
11/26/46
11/12/46
1.27,1.27.1.20
p t i m a t e d average partiole diameter (mioroscope), 30 or 35 @cmns. Line content of Doxvder obtained on small samples by weighing PYTOphosphate 98.80 98.82%. with estimated aoouraoy of -0.2%. Maxim& t e d p e r a t u r e of 88mple. 440' C. Residual powder gave x - r w diffiaotion pattern of rinc oxide only and liberated no hydrogen from acid.
ACKNOWLEDGMENT
The authors are indebted to W. F. Gilliam for suggesting the problem; to J. S. Kasper for the x-ray diffraction work: t o E. F. N l a m for establishing the particle size of the powder: and to A. C. Titus for determining its eiuc content. LITERATURE CITED
(1) Hansen, M., "Aufbau der Zweistofflegierungen,"p. 655, Berlin. Julius Springer, 1936. (2) Kel,ey, K. K,, Bur, Mines, Bull.383, 114(1935), (3) Price, J. W., J . SOC.C h m . Ind.. 45,283 (1945). (4) Vernon, w.H.. Akeroyd. E. I.. and Stroud. E. G . , J . I n s t . Metals. 45, No. 2, 301, (1939). RECBIYED
~ p t i 1 9 1949. ,
